El SIS en un proceso de trabajo biologista.

We conducted field research on Pinion Point Plantation (PPP) in Brooks County in the Red Hills

region of southern Georgia (2005–2007) and on Mount Pleasant Plantation (MPP) near Andrews, South Carolina (2008–2010). PPP consisted of 1,821 ha of ‘old field’ pine (Pinus spp.) forests with low basal areas (3–9 BA m2/ha) (72%), rotational fallow fields (10%), longleaf pine (Pinus palustris) plantings (5%), and bald cypress (Taxodium distichum) dome wetlands (13%). PPP was an established wild bobwhite property, but MPP was a new property developed beginning in 2005. Soils on MPP are primarily hydric and vegetation consists of dense stands of warm season grasses. Fallow fields, 1 to 2 ha in size, occur on 20% of property and are managed by annual disking. Intensive management for bobwhites at both sites includes prescribed burning, hardwood reduction, timber thinning, seasonal disking, drum chopping, nest predator reduction, and supplemental feeding. Both MPP and PPP have wild bobwhite populations with densities . 2.5 bobwhites/ha based on covey-call counts and hunting records.

METHODS

Outdoor Rearing Pens for Parent-reared Chicks

We constructed 16 trapezoidal-shaped rearing pens adjacent to one another (Stoddard 1931). The long sides of each pen were 5 m and the ends were 1 m on the short end and 3.6 m on the long end. Each pen had a 1-m2 shelter attached to the outside of each pen to facilitate changing water and providing feed. Pens were 2 m tall. The tops of the pens consisted of netting to allow for chick acclimation to local weather. Pens were enclosed by a snake fence and a solar-powered electric fence to exclude mammalian and reptilian predators. Vegetation in pens included common weeds (e.g., Ambrosia spp., Cassia spp.) to simulate natural brood habitat. A commercial operation was developed nearby beginning in 2009 following the same procedures and same source of eggs. This produced chicks for the studies on MPP; pens at Tall Timbers Research Station produced chicks for the PPP studies.

Rearing of Bobwhites

We obtained wild-strain bobwhite eggs from deserted nests on Tall Timbers Research Station which has maintained a wild bobwhite population for more than a century. Wild-strain eggs were hatched in an incubator and chicks were reared in brooders. These birds were used as layers to obtain eggs for the study. All chicks released were one generation removed from wild bobwhites with both parents from the wild. The wild-strain bobwhite chicks were raised either in a communal brooder (brooder-reared), communally-reared in flight pens (fall- released), or with parental imprinting and adoption (parent-reared).

Brooder-reared.—Brooder-reared chicks were re- moved from the incubator at hatching, placed into universal box type brooder pens (G. Q. F. Manufacturing Co., Savannah, GA, USA), and reared to 35 days of age. Brooder-reared chicks received commercial gamebird starter feed (Purina, St. Louis, MO, USA) with free-

standing waterers. Proso millet was mixed into the commercial feed at 2 weeks of age and grain sorghum was mixed into commercial feed at 4 weeks of age. Brooder-reared chicks were weighed at 12-days-of-age and numbered bands were attached to their right wing patagium (National Band and Tag Co., Newport, KY, USA). Brooder heaters were reduced at 21 to 35-days-of- age to prepare chicks for ambient temperatures upon release.

Parent-reared.—We played a recording of the calls hens produce on the nest when their eggs are hatching~ 36 hrs prior to hatching. This call series was previously recorded by placing a recording microphone in the clutch of wild bobwhite nests. Pre-hatch audio stimulation from parents has been found to have behavioral consequences in several bird species (Lickliter 2005). Chicks were taken from the incubator within 6 hrs of hatching and introduced to a wild-strain, bobwhite foster parent. We first placed foster parents in adoption boxes for 10–15 min after which 20 chicks were added behind a plexiglass divider. The divider was removed if the foster parent remained calm so chicks and adult came into contact and began the imprinting and adoption process. We removed the parent if a brood was rejected and added another potential parent. Parents of successfully-adopted chicks brooded and vocalized with chicks. We held adopted chicks and foster birds in a brooding box overnight in an attempt to strengthen their bond (Stoddard 1931). We released the brood with parent into the rearing pens the following morning where they remained for 35 to 42 days until release. No supplemental heating was provided. Chicks were fed, watered, and banded as described for pen-reared birds. All care, housing, and capture of bobwhites were in compliance with requirements of the University of Georgia’s Institutional Animal Care and Use Committee (AUP # A3437-01).

Fall-released Bobwhites.—This group of bobwhites was reared by a cooperating game bird breeder (Quail Call Farms, Beachton, FL, USA). Bobwhites were reared communally in the same age groups with minimal exposure to people and no exposure to adult bobwhites. Chicks were maintained in a heated brooder room attached to flight pens for 5 weeks and then released into flight pens until 10-weeks-of-age. Fall-released bobwhites were released on a separate section of the PPP property in fall 2005 and 2006 to avoid influencing the wild bobwhites and parent-reared bobwhites on our study area. Birds were banded the day before release with size 7 aluminum leg bands (National Band and Tag Company, Newport, KY, USA).

Study 1. Survival of Brooder-reared and Parent-reared Chicks

We released broods during 3 monthly periods during 2005 and 2006. Quail in 2005 were released in July, August, and September while in 2006 quail were released in July, August, and October. Release locations were selected based on known use of the area by wild bobwhite broods, and sites were recorded with global positioning systems (Trimble XT, Sunnyvale, CA, USA) for import-

ing into ArcGIS 9.2 (ESRI 2009). Release locations were 200 m apart to avoid mixing of parent-reared and brooder-reared chicks. The actual release site for each group was randomly selected and paired for each group (brooder- and parent-reared). Grain sorghum was broad- cast around the release location before the release and a recording of a bobwhite hen call played over a speaker to attract males or possibly females to the area to facilitate mixing with wild bobwhites for brood amalgamation (Faircloth et al. 2005). Quail were left with the release boxes which were removed the following day to minimize chick disturbance and provide shelter if needed.

Post-release Monitoring.—Recapture sessions were conducted in October and March following releases of parent-reared and brooder-reared bobwhites. October trapping sessions were ~ 2 weeks in length and the March trapping periods were~ 4 weeks. Bobwhites were captured using walk-in funnel traps baited with commer- cial grain sorghum (Stoddard 1931). Captured quail were classified to age and sex, weighed, and banded (if weighing  120 g) with an aluminum number 7 leg band. All birds were released at trap locations.

Parent-reared Survival Estimate.—We used Program MARK (White and Burnham 1999) to estimate survival of parent-reared bobwhites. We used the Burnham model incorporating live and dead recoveries (Palmer and Wellendorf 2007, Terhune et al. 2007) to calculate survival (u), recapture (q), and recovery (r) estimates (White and Burnham 1999). We used 5 intervals (LD-LD- LD-LD-LD) where the first 3 intervals were release periods and the following were October and March trapping sessions, respectively. Each interval designated as L included live trapping encounters, and the D interval included dead recoveries or encounters outside the designated trapping sessions. We defined interval lengths from July to August, August to September, September to October, and October to March for 2005 and 2006. Interval lengths in days for 2005 releases were 39, 39, 15, and 133, respectively, and interval lengths in days for 2006 releases were 39, 39, 15, and 128, respectively. We also modeled year as a covariant to assess variation in annual survival. We imposed several constraints prior to the analysis: recapture probabilities (q) for periods 1 and 2 were constrained to 0 because recapture (no trapping occurred) was not possible during these periods; recovery periods (r) 1, 2, and 3 were constrained to 0 because these were designated release periods and no harvest (recovery) occurred providing 0 probability of recovery; we constrained site fidelity (F) to 0.99, because radiotelem- etry revealed that movement from the study site was minimal. We used information-theoretic approaches to evaluate our biologically-derived candidate models (Burnham and Anderson 1998, Anderson et al. 2000). We used QAICc to compare the set of candidate models and considered the best model to have the lowest QAICc value (Burnham and Anderson 1998). We increased the precision of our estimates and accounted for model uncertainty by averaging parameter estimates over all candidate models that included the parameter of interest (Burnham and Anderson 1998, White and Burnham 1999).

Study 2. Reproductive Effort and Success on PPP

Parent-reared, brooder-reared, fall-released, and wild bobwhites weighing 140 g during March capture efforts received a 6-g (150–151 MHz), pendant-style, radio transmitter (American Wildlife Enterprises, Monticello, FL, USA) to monitor breeding season demographics. All birds were released at trap locations. Daily monitoring began in mid-April to document breeding season demographics and continued through September. Weekly monitoring began after this time period until the next breeding season. We monitored radio-marked quail  5 days a week. We flagged supposed nest sites when we found a bobwhite at the same location over a 2-day- period. We recorded clutch size when the incubating bird was on recess from incubation. We captured broods of radio-marked bobwhites at 8 days-of-age (Smith et al. 2003).

Data Analysis.—We report nests and broods per hen, nesting success, and brood survival for radio-marked hens. We based nests per hen and broods per hen on the number of radio-marked hens alive on 15 April. Nests per hen was the total number of nests divided by available radio-marked hens. Nest success was the proportion of nests that hatched  1 egg. Broods per hen was the number of successful hens hatching 1 egg divided by the number of radio-marked hens alive on 15 April. Apparent chick survival was based on calculations of chick survival rates for each brood and obtaining an average chick survival for all broods. We assumed there would be no differences among broods in rates of brood amalgamation (Faircloth et al. 2005).

Study 3. Reproductive Effort and Success on MPP

Parent-reared chicks were released on MPP during July through August 2008 and 2009 following protocols established for Studies 1 and 2. Chicks were recaptured the March following the year of release and a sample of wild and parent-reared bobwhites were banded and radio- marked as in Studies 1 and 2. We monitored summer survival and nesting activity of radio-marked bobwhites. We did not capture broods and band chicks in this study, but conducted flush counts at 3 weeks to compare brood- rearing success between wild and parent-reared bob- whites.

We calculated summer survival (1 Apr to 30 Sep) for resident wild and parent-reared, radio-marked bobwhites using Kaplan-Meier staggered entry (Pollock et al. 1989). We recognized sample sizes were low for Kaplan-Meier estimates and viewed these estimates with caution.

However, low sample size is more likely to bias survival rates lower than higher (Pollock et al. 1989, Palmer and Wellendorf 2007) and may be considered conservative for the purpose of estimating parent-reared bobwhite summer survival.

RESULTS

Study 1. Parent-reared Chick Survival on PPP

We released 595 chicks for each treatment over the 2- year study. We released 58, 45, and 50 parent-reared and brooder-reared bobwhites each in July, September, and August, respectively. Releases in 2006 for parent-reared and brooder-reared bobwhites each were 165, 96, and 181 for July, September, and October, respectively. We recaptured 68 parent-reared chicks in October and 35 in March; 3 were recovered during hunting. One of the brooder-reared chicks was recaptured in October and 5 were recaptured in March; 1 was recovered during hunting.

Low recapture of brooder-reared chicks (n ¼ 6) precluded estimating their survival rates. Parent-reared bobwhites had sufficient recaptures to estimate survival, recapture, and recovery probabilities. The model that minimized QAICc (u.p.r.f.) included the parameters survival, recapture, recovery, and site fidelity (fixed) being constant (Table 1). Model weight for the QAICc lowest model (wi¼ 0.41) provided evidence this was the

top model. The second best fitting model (uyearp.r.f.)

included year dependence for survival, but all other parameters were constant. Model weight for this model was close to the top model (wi¼ 0.30) and was 1.3 times

less likely than the model that minimized QAICc. Survival estimates for parent-reared bobwhites re- leased in July, August, and September 2005 until the October trapping session were 24.7, 44.4, and 79.8%, respectively (Table 2) and 42.6, 60.9, and 87.1%, respectively in 2006. Survival estimates for parent-reared bobwhites released in July, August, and September 2005 until the following March trapping session were 3.5, 6.2, and 11.1% , respectively (Table 3) and 12.8, 18.3, and 26.2% , respectively in 2006. Over-winter survival (Oct to Mar) estimates of parent-reared bobwhites was 14.0% and 30.1% for 2005 and 2006 releases, respectively.

Study 2. Reproductive Effort and Success on PPP

Wild (n¼ 35), parent-reared (n ¼ 7), and fall-released (n¼ 14) hens during 2006 incubated 37, 2, and 2 nests, respectively. Bobwhites successfully hatched 26, 1, and 1

Table 1. Survival (u), recapture (p), recovery (r), and fidelity (f) models for parent-reared bobwhite chicks tested for year effect (year) at Pinion Point Plantation in south Georgia during 2005–2007.

Models QAICc DQAICc Parameters Deviance wi

u.q.r.f. 278.494 0 3 272.460 0.40883

uyearq.r.f. 279.101 0.6074 4 271.044 0.30175

uyearqyearryearf. 280.150 1.6557 6 268.029 0.17865

uyearqyearr.f. 281.106 2.6116 5 271.019 0.11077

of these nests, respectively. Nests per hen was 1.06 (n¼ 37) for wild, 0.29 (n¼ 2) for parent-reared, and 0.14 (n ¼ 2) for fall-released bobwhites. Apparent nest success was 0.83, 0.50, and 0.50 for wild, parent-reared, and fall- released bobwhites in 2006, respectively. Average clutch sizes were similar 12.5 (11.5-13.6, 95% CI), 13.0 (9.1– 16.9, 95% CI), and 12.5 (9.6-15.4, 95% CI) for wild, parent-reared, and fall-released bobwhites. One female fall-released bobwhite had a clutch size of 27 eggs of which 21 hatched but had no chicks at 8-days-of-age. Apparent chick survival of wild, parent-reared, and fall- released bobwhite broods was 24.0% (n¼ 16), 71.0% (n ¼ 1), and, 0.0% (n¼ 1), respectively.

Wild (n ¼ 29), parent-reared (n ¼ 12), and fall- released (n¼ 21) hens in 2007 incubated 31, 4, and 19 clutches and successfully hatched 19, 3, and 8, respec- tively. Nests per hen was 1.07 (n¼ 31) for wild, 0.33 (n ¼ 4) for parent-reared, and 0.90 (n¼ 19) for fall-released bobwhites. Average clutch sizes were 14.4 (12.8-15.9, 95% CI), 14.3 (10.6-17.9, 95% CI), and 17.1 (15.3-19.0, 95% CI) for wild, parent-reared, and fall-released bobwhites in 2007, respectively. Parent-reared males incubated 4 nests in 2007 that were not included in the nests per hen statistic. Males hatched all 4 clutches and 2 broods were captured at 8-days-of-age. Apparent chick survival for wild, parent-reared, and fall-released bob- whites for the 2007 breeding season was 31.0% (n¼ 13), 28.0% (n¼ 6), and 13.0% (n ¼ 8), respectively.

Study 3. Reproductive Effort and Success on MPP

We released 843 and 2,345 parent-reared chicks on MPP during July-September 2008 and 2009. We radio- marked 27 parent-reared and 22 wild bobwhites in March and April. Summer Kaplan-Meier survival was 0.39 6 0.18 for wild bobwhites and 0.276 0.12 for parent-reared bobwhites. We monitored reproductive success of parent- reared hens (n¼ 11) and wild resident hens (n ¼ 19) on MPP. Parent-reared hens produced 7 nests and hatched 6 broods resulting in nests per hen and broods per hen of 0.64 and 0.55, respectively. Clutch size averaged 14.36 1.76 eggs. Wild bobwhites produced 10 nests and hatched 6 clutches resulting in nests per hen at 0.53 and hatches per hen at 0.32. Clutch size averaged 12.2 6 0.84 eggs. The proportion of hens with broods at 3 weeks was 20% for both wild and parent-reared bobwhites.

We radiomarked 25 wild bobwhites and 31 parent- reared bobwhites in 2010. Summer Kaplan-Meier survival was 0.3876 0.119 and 0.295 6 0.101 for parent-reared bobwhites. We monitored reproductive success of parent- reared hens (n¼ 13) and wild resident hens (n ¼ 25) on MPP. Parent-reared hens produced 9 nests and hatched 4 clutches resulting in nests per hen and brood per hen of 0.69 and 0.31, respectively. Clutch size averaged 10.456 1.06 eggs. Wild bobwhites produced 18 nests and hatched 9 clutches resulting in nests per hen of 0.36 and broods per hen of 0.31. Clutch size averaged 11.86 1.01 eggs. Fifty percent of parent-reared hens had broods at 3 weeks (2 of 4 hens) compared to 33% of wild bobwhites (3 of 9 hens).

Sixteen nests were produced by 24 parent-reared hens and 28 nests were produced by 44 wild hens over both years. Nests per hen was 66.7% for parent-reared and 63.6% for wild bobwhites. Ten broods were produced by the parent-reared bobwhites and 15 broods were produced by wild hens. Broods per hen was 41.7% for parent-reared and 34.1% for wild bobwhites.

DISCUSSION

Study 1. Survival of Released Chicks

Over-winter survival estimates of parent-reared bobwhites were higher than previous survival estimates of released pen-reared bobwhites. Pierce (1951) reported pen-reared bobwhite over-winter survival of 7% whereas DeVos and Speake (1995) reported 20% survival of pen- reared bobwhites to April. Perez et al. (2002) reported no survival of released bobwhites. Brooder-reared chicks in our study had low survival rates unlike parent-reared chicks. This was likely a combination of improper rearing and lack of imprinting. Ring-necked pheasants (Phasianus colchicus) reared with species-specific foster parents had greater clutch and brood survival than brooder-reared counterparts (Brittas et al. 1992). We noticed that when chicks imprinted to a species-specific parent, they immediately developed fear of humans and were more likely to express normal predator avoidance behaviors (Dowell 1992). Parent-reared chicks rarely gave lost calls; whereas, brooder-reared chicks frequently gave lost calls

Table 2. Parent-reared bobwhite modeled averaged survival estimates and confidence intervals from release to fall trapping sessions, Pinion Point Plantation, Brooks County, Florida, 2005– 2006.

95% CI Periods Year Estimate LCI UCI Jul-Oct 2005 0.247 0.0333 0.5652 2006 0.426 0.3080 0.5388 Aug-Oct 2005 0.444 0.1386 0.718 2006 0.609 0.5047 0.6983 Sep-Oct 2005 0.798 0.5776 0.9121 2006 0.871 0.8270 0.9051

Table 3. Parent-reared bobwhite modeled averaged survival estimates and confidence intervals on Pinion Point Plantation from release to capture in March 2005 and 2006, Brooks County, Florida.

95% CI Periods Year Estimate LCI UCI Jul-Mar 2005 0.035 0.0003 0.253 2006 0.128 0.0586 0.2255 Aug-Mar 2005 0.062 0.0011 0.3214 2006 0.183 0.0961 0.2923 Sep-Mar 2005 0.111 0.0048 0.4083 2006 0.262 0.1574 0.3788 Oct-Mar 2005 0.140 0.0083 0.4476 2006 0.300 0.1904 0.4185

which also likely influenced survival. Parent-reared chicks were more often captured with wild bobwhites; whereas, brooder-reared bobwhites were often found in like groups. Brooder-reared chicks were also less fearful of humans and were less likely to hide in vegetation than parent-reared bobwhite chicks.

Our survival estimates suggest a demographically- significant number of parent-reared chicks could survive until the following breeding season and positively influence future populations if recruitment was adequate. The combination of imprinting to species-specific foster parents and rearing in a semi-natural environment may have increased chick survival through appropriate behav- ioral responses (Roseberry et al. 1987, Dowell 1992). Wild bobwhites may have also had a role in increased survival through adoption of chicks by wild bobwhite broods (Faircloth et al. 2005).

Survival estimates for parent-reared bobwhites re- leased in 2006 were higher than in 2005 possibly because of different weather and habitat conditions. Rainfall amounts were greater during summer months from tropical storms and hurricanes in 2005, which likely reduced chick survival. Thinning of timber stands and intensive dragging of steel tracks over burned areas to increase soil disturbance resulted in thin cover and possibly attributed to increased winter mortality. Pre- scribed burning was conducted using recommended management guidelines in 2006 (Masters et al. 2003) and other habitat manipulations were minimal on study areas.

Study 2. Reproductive Effort and Success on PPP

Sample sizes were low for parent-reared chicks and likely an artifact of sampling. For example, based on our survival estimates and releases scattered across the study area, ~150 chicks should have survived until the March trapping session over the 2 years. We recaptured~23% of these in March. Our sample size of hens were small given